Steam Temperature Prediction Control Method Using Prediction Model of Multi Heat Exchangers for Thermal Power Plants

Abstract

This paper proposes a new prediction control method for thermal power plants. In the method, steam temperatures are predicted using prediction model of multi heat exchangers. Each model of a heat exchanger consists of a lumped parameter physical model and a dead time element. State variables of the heat exchanger are estimated using a Kalman Filter, and near future changes of steam temperatures are predicted by integrating multi heat exchanger models, and fuel flow and water spray flows are compensated accordingly.
The results of simulation using a practical scale simulator of the thermal power plant show that the proposed method is so practical that changes of main steam temperature are reduced by half comparing with those of PID control method.